A hybrid calibration scheme for developing hydrogen enrichment ratio control map using RSM and ANN technique to enhance the characteristics of an ammonia biodiesel RCCI combustion engine

Abstract

This investigation aims to optimally attune hydrogen enrichment ratio for an ammonia biodiesel powered RCCI engine using RSM and ANN techniques. Conventionally, optimal mapping is accomplished through trial-and-error experimentation. In this study, an innovative model-based calibration method for developing an optimal hydrogen ratio map is suggested. The LRF energy share of ammonia in premixing is set at 40%, the hydrogen enrichment is varied from 5 to 20%, and the rest of the energy is biodiesel as HRF. Based on the experimental inputs, the RSM-based optimization has established an ideal hydrogen enrichment ratio of 14.77%, 18.75%, 18.93%, 17.94%, 16.43%, 14.42%, 15.58%, 13.23%, and 17.12% for intermittent load conditions of 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 100%. A simultaneous decrease in smoke and BSEC is obtained, with a little penalty in NOx emissions at optimal conditions. RSM optimization with a desirability level of 97.9% was conducted over three trials for authentication using ANN. Based on the results, ANN predicts all the replies with R > 0.96. The experimental corroboration of the predicted variables of the optimized map for 50% load has an error between 1.40% and 4.95%, which are within the acceptable range.